An Elemental Impurities Excipient Database: A Viable Tool for ICH Q3D Drug Product Risk Assessment.

To support the practical implementation of the International Council for Harmonisation (ICH) Q3D guideline, which describes a risk-based approach to the control of elemental impurities in drug products, a consortium of pharmaceutical companies has established a database to collate the results of analytical studies of the levels of elemental impurities within pharmaceutical excipients. This database currently includes the results of 26,723 elemental determinations for 201 excipients and represents the largest known, and still rapidly expanding, collection of data of this type. Analysis of the database indicates good coverage of excipients relevant to real-world drug product formulations and tested element profiles consistent with ICH Q3D recommendations. The database includes the results from multiple analytical studies for an excipient and thus incorporates within it an indication of both excipient supplier and batch-to-batch variability as well as any variability associated with the different testing organizations and methods employed. The data confirm the findings of earlier smaller studies that elemental impurity concentrations in excipients are generally low and when used in typical proportions in formulated drug products are unlikely to pose a significant patient safety risk. The database is now in active use as one line of evidence in ICH Q3D risk assessments.

A k-Nearest Neighbours Approach Using Metabolism-related Fingerprints to Improve In Silico Metabolite Ranking.

The application of biotransformation dictionaries derived by expert evaluation of known metabolic pathways represents one approach to the prediction of both phase I and phase II xenobiotic metabolites. The ranking of metabolites generated by such dictionaries has previously been achieved through the use of qualitative reasoning rules and quantitative probability values. Using the biotransformation dictionary available in the Meteor expert system, we show that metabolite over-prediction by both of these methods can be reduced by the adoption of a k-nearest neighbours methodology in which the likelihood of a predicted biotransformation is determined based on comparison of a query chemical with structurally-similar substrates with known experimental metabolic data which activate the same biotransformation. Optimal performance was achieved when similarity was defined in terms of a combination of two fingerprints, one describing the overall profile of biotransformations a structure can potentially undergo and the other describing the local environment around the predicted site of metabolism for the particular biotransformation under consideration.

An expert system approach to the assessment of hepatotoxic potential.

Hepatotoxicity is a major cause of pharmaceutical drug attrition and is also a concern within other chemical industries. In silico approaches to the prediction of hepatotoxicity are an important tool in the early identification of adverse effects in the liver associated with exposure to a chemical. Here, we describe work in progress to develop an expert system approach to the prediction of hepatotoxicity, focussing particularly on the identification of structural alerts associated with its occurrence. The development of 74 such structural alerts based on public-domain literature and proprietary data sets is described. Evaluation results indicate that, whilst these structural alerts are effective in identifying the hepatotoxicity of many chemicals, further research is needed to develop additional structural alerts to account for the hepatotoxicity of a number of chemicals which is not currently predicted. Preliminary results also suggest that the specificity of the structural alerts may be improved by the combined use of applicability domains based on physicochemical properties such as log P and molecular weight. In the longer term, the performance of predictive models is likely to benefit from the further integration of diverse data and prediction model types.

Identification of structure-activity relationships for adverse effects of pharmaceuticals in humans: Part C: use of QSAR and an expert system for the estimation of the mechanism of action of drug-induced hepatobiliary and urinary tract toxicities.

This report describes an in silico methodology to predict off-target pharmacologic activities and plausible mechanisms of action (MOAs) associated with serious and unexpected hepatobiliary and urinary tract adverse effects in human patients. The investigation used a database of 8,316,673 adverse event (AE) reports observed after drugs had been marketed and AEs noted in the published literature that were linked to 2124 chemical structures and 1851 approved clinical indications. The Integrity database of drug patent and literature studies was used to find pharmacologic targets and proposed clinical indications. BioEpisteme QSAR software was used to predict possible molecular targets of drug molecules and Derek for Windows expert system software to predict chemical structural alerts and plausible MOAs for the AEs. AEs were clustered into five types of liver injury: liver enzyme disorders, cytotoxic injury, cholestasis and jaundice, bile duct disorders, and gall bladder disorders, and six types of urinary tract injury: acute renal disorders, nephropathies, bladder disorders, kidney function tests, blood in urine, and urolithiasis. Results showed that drug-related AEs were highly correlated with: (1) known drug class warnings, (2) predicted off-target activities of the drugs, and (3) a specific subset of clinical indications for which the drug may or may not have been prescribed.

In silico tools for sharing data and knowledge on toxicity and metabolism: derek for windows, meteor, and vitic.

ABSTRACT Lhasa Limited is a not-for-profit organization that exists to promote the sharing of data and knowledge in chemistry and the life sciences. It has developed the software tools Derek for Windows, Meteor, and Vitic to facilitate such sharing. Derek for Windows and Meteor are knowledge-based expert systems that predict the toxicity and metabolism of a chemical, respectively. Vitic is a chemically intelligent toxicity database. An overview of each software system is provided along with examples of the sharing of data and knowledge in the context of their development. These examples include illustrations of (1) the use of data entry and editing tools for the sharing of data and knowledge within organizations; (2) the use of proprietary data to develop nonconfidential knowledge that can be shared between organizations; (3) the use of shared expert knowledge to refine predictions; (4) the sharing of proprietary data between organizations through the formation of data-sharing groups; and (5) the use of proprietary data to validate predictions. Sharing of chemical toxicity and metabolism data and knowledge in this way offers a number of benefits including the possibilities of faster scientific progress and reductions in the use of animals in testing. Maximizing the accessibility of data also becomes increasingly crucial as in silico systems move toward the prediction of more complex phenomena for which limited data are available.

Combined Use of MC4PC, MDL-QSAR, BioEpisteme, Leadscope PDM, and Derek for Windows Software to Achieve High-Performance, High-Confidence, Mode of Action-Based Predictions of Chemical Carcinogenesis in Rodents.

ABSTRACT This report describes a coordinated use of four quantitative structure-activity relationship (QSAR) programs and an expert knowledge base system to predict the occurrence and the mode of action of chemical carcinogenesis in rodents. QSAR models were based upon a weight-of-evidence paradigm of carcinogenic activity that was linked to chemical structures (n = 1,572). Identical training data sets were configured for four QSAR programs (MC4PC, MDL-QSAR, BioEpisteme, and Leadscope PDM), and QSAR models were constructed for the male rat, female rat, composite rat, male mouse, female mouse, composite mouse, and rodent composite endpoints. Model predictions were adjusted to favor high specificity (>80%). Performance was shown to be affected by the method used to score carcinogenicity study findings and the ratio of the number of active to inactive chemicals in the QSAR training data set. Results demonstrated that the four QSAR programs were complementary, each detecting different profiles of carcinogens. Accepting any positive prediction from two programs showed better overall performance than either of the single programs alone; specificity, sensitivity, and Chi-square values were 72.9%, 65.9%, and 223, respectively, compared to 84.5%, 45.8%, and 151. Accepting only consensus-positive predictions using any two programs had the best overall performance and higher confidence; specificity, sensitivity, and Chi-square values were 85.3%, 57.5%, and 287, respectively. Specific examples are provided to demonstrate that consensus-positive predictions of carcinogenicity by two QSAR programs identified both genotoxic and nongenotoxic carcinogens and that they detected 98.7% of the carcinogens linked in this study to Derek for Windows defined modes of action.

Structure-activity relationships for skin sensitization: recent improvements to Derek for Windows.

Derek for Windows (DfW) is a knowledge-based expert system that predicts the toxicity of a chemical from its structure. Its predictions are based in part on alerts that describe structural features or toxicophores associated with toxicity. Recently, improvements have been made to skin sensitization alerts within the DfW knowledge base in collaboration with Unilever. These include modifications to the alerts describing the skin sensitization potential of aldehydes, 1,2-diketones, and isothiazolinones and consist of enhancements to the toxicophore definition, the mechanistic classification, and the extent of supporting evidence provided. The outcomes from this collaboration demonstrate the importance of updating and refining computer models for the prediction of skin sensitization as new information from experimental and theoretical studies becomes available.

Using argumentation for absolute reasoning about the potential toxicity of chemicals.

The application of a new argumentation model is illustrated by reference to DEREK for Windows, a knowledge-based expert system for the prediction of the toxicity of chemicals. Examples demonstrate various aspects of the model such as the undercutting of arguments, the resolution of multiple arguments about the same proposition, and the propagation of arguments along a chain of reasoning.